轉換資料型別

SQL 報表製作

Tyler Pernes

Learning & Development Consultant

本章目標

「雜亂」的資料型別

議題 1：型別限定的函式

議題 2：合併資料表（JOIN）

議題 2：合併資料表（UNION）

解讀錯誤訊息

型別限定函式錯誤：

SELECT AVG(first_name)
FROM athletes;

ERROR: Function avg(character varying) does not exist

JOIN 錯誤：

SELECT country, continent
FROM countries AS c1
JOIN continents AS c2
ON c1.continent_id = c2.id;

ERROR: Operator does not exist: integer = character varying

解法：用 CAST() 包起來

語法：

CAST(field AS type)

範例：

CAST(birthday AS date)
CAST(country_id AS int)

為函式做 CAST

SELECT DATE_PART('month',birthdate)
FROM birthdates;

ERROR: Can't run DATE_PART on string.

SELECT DATE_PART('month', 
       CAST(birthdate AS date)) 
FROM birthdates;

+-----+
| 04  |
| 05  |
+-----+

為 JOIN 做 CAST

SELECT a.id, b.id
FROM table_a AS a
JOIN table_b AS b
ON a.id = b.id;

ERROR: Cannot join ON varchar = int.

SELECT a.id, b.id
FROM table_a AS a
JOIN table_b AS b
ON a.id = CAST(b.id AS varchar);

Query Ran Successfully!

面對資料型別問題的規劃

問題出現就修正
讀懂錯誤訊息！

面對資料型別問題的規劃

SELECT column_name, data_type
FROM information_schema.columns
WHERE table_name = 'countries';

+--------------+--------------------+
| column_name  | data_type          |
|--------------|--------------------|
| id           | integer            |
| country      | character varying  |
| region       | character varying  |
+--------------+--------------------+

資料型別文件

https://www.postgresql.org/docs/9.5/datatype.html

Practice time!

SQL 報表製作